bangor business school working paper€¦ · payday loans or cash advances, are structured to...
TRANSCRIPT
Bangor Business School
Working Paper (Cla 24)
BBSWP/12/008
Can an interest-free credit facility be more efficient than a usurious
payday loan?
(Bold, Clarendon 12, capitals)
By
Murizah Osman Salleh
Bank Negara Malaysia and Bangor University
Aziz Jaafar
Bangor University
M. Shahid Ebrahim
Bangor University
July 2005
Bangor Business School
Bangor University
Hen Goleg
College Road
Bangor
Gwynedd LL57 2DG
United Kingdom
Tel: +44 (0) 1248 382277
E-mail: [email protected]
2
Can an interest-free credit facility be more efficient than a usurious payday loan?
Murizah Osman Salleh
Bank Negara Malaysia and Bangor University
Aziz Jaafar
Bangor University
M. Shahid Ebrahim
Bangor University
This draft: July 5, 2012
Acknowledgments: We have benefitted from the critical comments of John Goddard and
Jeremy Tobacman on earlier drafts of the paper. Any remaining errors are due to our
negligence.
Correspondence Address: Murizah Osman Salleh
Bangor Business School
College Road, Hen Goleg
Bangor, Gwynedd LL57 2DG
United Kingdom
E-Mail: [email protected]
Can an interest-free credit facility be more efficient than a usurious payday loan?
Abstract: Permanent disequilibrium in mainstream credit markets have pushed the unbanked
and underbanked households to frequent high cost payday loans for their liquidity needs.
Associated with the latter are welfare-reducing issues of predation and debt-entrapment. In
response to this market failure, we expound a simple model that integrates inexpensive
interest-free liquidity facility within an endogenous leverage circuit. This builds on the
technology of ROSCA/ ASCRA/ mutual/ financial cooperative and cultural beliefs
indoctrinated in Islam. Results indicate that such a circuit moderates adverse selection and
moral hazard issues more efficiently than payday loan and mainstream financier.
Additionally, it does not suffer the drawbacks of welfare-reducing payday loans and also
addresses financial exclusion in mainstream credit markets.
JEL codes: D14, G29, G32, 016, Z12
Keywords: interest-free loan, payday loan, financial exclusion, liquidity facility, cooperatives
I. Introduction
“Financial exclusion is likely to act as a “brake” on development as it retards
economic growth and increases poverty and inequality. .... Without access to financial
services, households will not be able to invest in their human capital, smooth consumption,
insure against income shocks, and save for retirement”.
Beck and Demirguc-Kunt (2007, p.1)
The Federal Deposit Insurance Corporation (FDIC) 2009 national survey of unbanked
and underbanked households carries concerns on the extent of financial exclusion of
American households.1 Approximately 26% or 38 million households, consisting of 60
million adults, mostly residing in urban and rural areas, are financially excluded from
mainstream credit markets. Clearly, as articulated above by Beck and Demirguc-Kunt (2007),
this has negative repercussions on the households‟ ability to achieve financial security,
impedes economic growth and creates social imbalances.
The credit market for the unbanked and underbanked household exhibits the classic
case of „permanent disequilibrium‟ discussed by Baltensperger (1978), caused by the twin
issues of adverse selection and moral hazard, which results in a persistent credit rationing, or
complete withdrawal of mainstream financiers from the market (Stiglitz and Weiss, 1981).
To bridge their credit needs, these households have had to frequent high cost alternative
financial service providers, or fringe banking institutions. Of particular interest to this paper
is the reliance on payday loans as a financial solution for emergency and expenditure shocks.
Payday loans or cash advances, are structured to function as a short term liquidity
facility to smooth inter-temporal income shocks. This involves issuance of single, small,
short-term and unsecured consumer loan, ranging from $100 to $500. An average payday
loan is for less than $300, with repayment period of 7 to 30 days (Elliehausen and Lawrence,
2001; Stegman, 2007). The industry has been severely criticised for its high credit cost, in
1 According to the FDIC (2009), unbanked households are those currently without any savings or checking
account. On the other hand, the underbanked are those who do have banking accounts but subscribe to the
services of alternative financial service providers. Based on this categorisation, 7.7% (9 million) and
17.9% (21 million) households are unbanked and underbanked, respectively. Additionally 5 million
households may potentially be underbanked but have been omitted from the above due to paucity of data
on their usage of alternative financial services.
combination with wider issues of predatory practices and expropriation of wealth.
2
Undergirding these criticisms is the fact that payday loans exacerbate debt entrapment on a
subset of its customers (Elliehausen and Lawrence, 2001), who are economically vulnerable
(Smith et al., 2008), and lack financial sophistication (Flannery and Samolyk, 2005). The
onerous fees reflects the industry‟s high transactional costs (Skiba and Tobacman, 2007), and
severe default rates (DeYoung and Phillips, 2009).3 Interestingly, despite heavy criticisms,
there is still persistent demand for the products. Thus, this highlights a pressing need to
explore inexpensive financial alternatives to assuage the liquidity needs of unbanked and
underbanked population.4 To date, studies on payday loans have either focused on (i) credit
behaviours; or (ii) welfare effect of the borrowers, without delving on Pareto optimal
substitutes.
Recognising this shortcoming, the primary motivation of this paper is to expound an
institutional design for the provision of inexpensive, short-term liquidity facility, which
satisfies the latent demand of this underserved market to smooth their inter-temporal
exogenous income shocks. Specifically, our study aims to explore the following question:
Can an endogenous interest-free payday loan circuit provide a more efficient credit solution
in contrast to current payday lenders and mainstream financiers? This is achieved through
integrating the two strands of literature on: (i) institutional structures related to endogenous
circuits (such as Rotating Savings and Credit Association – ROSCA/ Accumulating Savings
and Credit Associations – ASCRA); with (ii) cultural beliefs (i.e., Islamic tenets) in
particular, interest-free loans.5 Our research motivation is consistent with that of Coase
2 Predatory lending is “welfare reducing provision of credit” (Morgan, 2007, p.1), characterised by
“excessively high interest rates or fees, and abusive or unnecessary provisions that do not benefit the
borrower” (Carr and Kolluri, 2001, p.1).
3 These are mainly fixed overheads, loan processing, taxes, and funding costs (Ellihausen and Lawrence,
2001). The industry‟s default rate of 21% is extremely risky compared to the 3% rate experienced by
commercial banks (DeYoung and Phillips, 2009). We find that the high cost concurs with credit literature
to compensate for risk associated with these risky borrowers.
4 Although we have used the United States as the primary reference base, this does not preclude the
existence of payday lending in other developed and developing economies.
5 ROSCA and ASCRA are informal institutions, where members contribute periodically an amount of funds
to a common pool over a specified period. In ROSCA, the sequence of assignment of the pooled funds to
each member is either determined on a random basis or through a bidding process. By pooling resources,
it permits the mobilization of funds that otherwise would have been kept out of circulation (Besley et al.,
1994). Whilst ASCRA shares similar features of its nemesis, there is greater flexibility in the amount and
timing of each member contribution, larger membership, allocation of the pooled funds, and its greater
social function (Bouman, 1995).
(1937) and Alchian (1950), who in their seminal papers rationalise efficient institutions as
those that evolve and adapt to the environment to deliver services in a cost effective manner.
Moreover, the approach taken in this paper to intertwine institutional design with culture is
reflective of Acemoglu et al. (2005, p.424), who reiterate “belief differences clearly do play a
role in shaping policies and institutions”.
For the purpose of this paper, the target population are economically active households.
This is consistent with the underwriting criteria of payday lenders that require borrowers to
be in employment, as well as with the findings of the FDIC (2009) survey. Additionally, our
model is based on risk neutral economic agents. We illustrate the above through an
institutional structure of an endogenous leverage circuit formed from member based
contributions.6 This is followed by two stepped extensions that assimilate real world
elements of having (non)borrowers within a finite life circuit, and subsequently extending the
circuit as a going concern with random repetitive (non)borrowing. The objective of the basic
framework and the extensions are to solve for Pareto-efficiency. This is showcased by
mathematically modelling a short term interest-free credit facility circuit that moderates
agency costs of debt. The beauty of the model lies in the structuring of the circuit, where
members help one another to alleviate inter-temporal liquidity shocks such that the benefits
of borrowing outweigh the cost of it. This draws from the „barn raising‟ practices in the
United States frontiers discussed in Besley et al. (1993) and captures Commons‟s viewpoint
(1931, p.651) where he states “... collective action is more than control and liberation of
individual action-it is expansion of the will of the individual far beyond what he can do by his
own puny acts”.7
This paper contributes to existing literature from four perspectives. First, it averts
expropriation of wealth of unbanked and underbanked households through establishing an
alternative recourse for liquidity funding. This is in contrast with liquidity stripping from
6 We employ a generic term „circuit‟ to signify all institutions where the principal and agent are the same
individual. The structure is akin to that of a non-profit institution. An administrator may be present but is
not incentivised by rent-seeking motives.
7 Our model reiterates the significant developmental role of endogenous circuits in the 19
th century. These
circuits permit greater latitude to grant its customers affordable credit compared to profit-oriented
mainstream financiers (El-Gamal, 2009). Recently, the economic importance of endogenous circuits in the
United Kingdom was further boosted by the legislative reforms that enabled these institutions (i) greater
market reach; and (ii) flexibility to determine its member incentive structures (see HM Treasury, 2012;
Moore, 2012).
onerous interest charges of current payday loans that self-reinforces the already poor
economic status of these households. Second, our framework allows the casting of financial
inclusion to also include those unbanked (i.e., households without savings or checking
accounts), who would normally be precluded by mainstream financiers. Third, we integrate
interest-free loans in our liquidity facility.8 This is drawn from charitable teachings,
specifically from Islamic religious tenets, that are proffered as a remedy to the prohibited
interest (ribā an-nasi‟ah).9 Thus, it unveils the economic potential of this antiquated
financing, conceived from cultural ideals, as a financial development device. Additionally, it
challenges the widely-held view of interest as the primary credit allocation mechanism.
Fourth, by binding eligibility to the liquidity facility with a member‟s fulfilment of the
periodical contributions ruling, it harnesses the commitment technology sacrosanct with
endogenous leveraged circuit-based institutions. This effectively moderates the issue of time-
inconsistent preferences closely associated with payday loan borrowers.
The remaining of this paper is structured as follows. Section II details the landscape of
the payday lending industry and related literature. Section III discusses the rationale for the
prohibition of interest and its contrast against charitable modes in Islamic tenets. In Section
IV, we develop a simple model to illustrate the Pareto-efficiency of endogenous interest-free
payday loan circuit in addressing financial autarky of unbanked and underbanked households
and its results. Finally, we conclude in Section V.
II. Landscape of payday industry and related literature
Payday lending emerged in early 1990s in response to increased demand for short term
credit following the spatial void created by withdrawal of mainstream banks from small
loans, low profit margin business segment (Flannery and Samolyk, 2005; Smith et al., 2008;
OFT, 2010). The convenience of fast disbursement, minimal or non-existent credit checks
further adds to its attractiveness (Caskey, 2005). An indicator of its growth pace is the
8 Here, we are drawn to the statement by the well-known Islamic scholar Fazlur Rahman (1964), who
compartmentalises this form of funding to only philanthropic endeavours. The probable reason behind this
limited worldview is because the Islamic jurisprudence (fiqh) literature is silent on the structuring of
interest-free loan (qard hasan) for assisting the underprivileged. Such shortcomings in the development of
fiqh provide ammunition to critique Islamic law (see Kuran, 2011).
9 Unsurprisingly, charitable concept of interest-free funding is also present in other Abrahamic faith. For
example, the existence of Jewish free loan societies is linked to the obligation in Judaism for extending
free loans to the poor (Lewinson, 1999).
extensiveness of payday loan network across the United States. Payday lenders have more
branch presence then McDonalds and Starbucks combined (Zinman, 2010). Based on the
2009 FDIC survey, approximately four million unbanked and underbanked households have
frequented payday lenders, which is now a $38.5 billion industry (FDIC, 2009; CFSA,
2011).10
Unlike those served by micro-financiers, payday loan customers are relatively
economically better off, for they must be employed and banked to subscribe to these services.
They also differ from pawn broking customers due to their contrastingly higher income
bracket (Johnson and Johnson, 1998), where majority of the borrowers enjoy moderate
earnings of $25,000 to $50,000. Subscribers are mostly young, below the age of 45, single
parent households, with middle level education (Elliehausen and Lawrence, 2001).11
This
category fits the profile of having lesser degree of financial sophistication presented in
consumer finance theories.
The survey by Ellihausen and Lawrence (2001) finds majority are infrequent users of
the payday lending facility. However, there are a selected few, accounting for 23% of total
customers that have 14 or more loans in the same year.12
These frequent users tend to roll
over the outstanding loan. Generally, these loans would run for 2 weeks or less, or over a 3-4
week periods. These chronic borrowers are more likely to have exposure to more than one
payday loan, exhibiting the classic case of borrowing from Peter to pay Paul, where a loan
drawn on a new payday lender is often used to offset against an old one. Payday loan
customers do subscribe to other means of credit, such as, bank and retail credit cards, and
auto loans. However, most have maximised their credit limits on these sources of finance.
Payday loan is attractive to its customers for the convenience and fast disbursement. Unlike
mainstream financiers, credit checks are minimal.
10
In the United Kingdom, it is estimated a total of 1.2 million adults have taken out payday loans, amounting
to £1.2 billion advances in 2009, and this is expected to rise further in the coming years. The United
Kingdom payday lending industry is highly concentrated. One of the seven major players controls 25% of
the 2009 total market share (Burton, 2010). However, this is expected to change with greater influx of
payday lenders into the market, particularly from the United States (Goff and O‟Connor, 2011).
11 The United Kingdom payday loan market is generally similar to the United States except for its customer
base and loan tenures. The borrowers are from a younger set under the age of 35, single, without children
and income of £25,000 per annum. The payday loan runs on monthly basis, averaging 1 – 2 months
(Burton, 2010).
12 This is similarly observed by Flannery and Samolyk (2005) in their sample of two leading payday lenders.
The primary complaints against payday lenders are the exorbitant finance charge. Fees
for a $100 loan ranges from $15 to $30, with annual percentage rate (APR) of 20-300%. The
extremely high cost in contrast to other near credit substitutes raises criticism from consumer
advocates and public agencies.13
According to industry players, the APR is resultant from the
small loan size, given payday lenders‟ fixed overhead costs coupled with the high default
rates. Industry players argue that the $15 charge is definitely lower than the $50 flat rate
returned check fees or a $25 covered overdraft (overdraft protection) by depository
institutions (Morgan et al., 2012).
Payday lending is a regulated industry. It is subjected to state and federal laws and
some players also subscribe to industry standards of the Community Finance Services
Association (CFSA); an industry self regulatory organisation (Elliehausen and Lawrence,
2001).14
The United States recently established a new regulatory body, the Consumer
Financial Protection Bureau in July 2011 to oversee matters related to consumer protection,
including market conduct of payday loan industry. This independent body is part of the
financial reforms outlined in the Dodd-Frank Wall Street Reform and Consumer Protection
Act 2010.15
Studies on payday lending have primarily centred on two aspects; namely consumer
credit behaviours and welfare effects of the availability or withdrawal of this credit. Skiba
and Tobacman (2008) seek to rationalise the demand for payday loans despite its excessively
high fees. They find that payday loan borrowers exhibit partially naive quasi hyperbolic
discounting tendencies. In that, the borrowers demonstrate overly optimistic forecast of future
outcomes in respect of their own time preference, or their probability of absorbing future
shocks.
13
Bertrand and Morse (2011) in testing effects of information disclosure on correcting cognitive biases,
provide the following contrasting APRs: (i) payday advance 443%, (ii) instalment car loans 18%, (iii)
credit card 16%, and (iv) subprime mortgages 10%.
14 The industry is subjected to: (i) the Truth in Lending Act at the federal level that governs disclosure
requirements; (ii) Fair Debt Collection Act that regulates debt collection practices; and (iii) National Bank
Act that essentially allows the payday lender to enter into rent-a-bank model, which is now defunct by
virtue of the stricter FDIC regulation on national chartered banks. The CFSA provides industry the best
practices that are essentially focused on consumer protection.
15 The governing act for payday loan in the United Kingdom is the Consumer Credit Act 1974 and the
industry is regulated by the Office of Fair Trading (OFT).
Agarwal et al. (2009) find that the sampled population choose payday loans despite
having unused liquidity on their credit cards. This exemplifies existence of liquid debt
puzzle, whereby individuals undervalue their financial options. This also highlights the
individuals‟ lack of cognitive ability to discern costs across different financial products.
Gathergood (2012) points persistent indebtedness to poor financial literacy and self-control
problems. In such a case, individuals are more likely to succumb to impulsive consumption.
The ease of credit provided by high-cost credit providers including payday loan, further
exacerbates this tendency and heighten the likelihood of over-indebtedness. Campbell et al.
(2010) warrant that any public policy decisions must adopt a balanced approach between
designing „behaviourally informed regulatory changes‟ to correct consumers‟ behavioural
tendencies, i.e., present-bias preferences and limited cognitive ability, with assessment on the
economic implications to the payday loan industry. This is aligned with Gathergood (2012),
who supports paternalistic approaches in regulations, i.e., preventing access to credit that
pushes consumers to succumb to sub-optimal behaviour.
In regards to its welfare effects, the evidence is still debatable, “Do payday lenders, on
net, exacerbate or assuage customers‟ financial difficulties?” (Caskey, 2010, p.38). Morse
(2011) and Zinman (2010) to name a few, argue that accessibility to payday loan is welfare
enhancing, which is in contrast to Melzer (2011) and Morgan et al. (2012).16
Using the 1996
natural disaster in California as an event that has widespread economic effect on households,
Morse (2011) finds that presence of payday lenders reduces emergency distress and serious
criminal incidences. Zinman (2010) finds that restricting access creates deterioration in
financial position of Oregon households as opposed to those domiciled in Washington, as the
control state. Households in the restricted payday state experienced higher unemployment
and reported an overall poor future financial outlook. The negative effect of the regulatory
ban on payday loan is worsened by the lack of financial substitutes.
III. Islamic prohibition of interest and the contrast against charity
16
Thirty three states permit payday lending with rules on payday loan terms including maximum fees,
rollovers, loan size, licensing and examination requirements as well as collection procedures for past-due
loans. Seventeen states totally prohibit offering of payday loans. Contrary to the United States, the United
Kingdom refrains from adopting intrusive regulatory measures, such as stringent price controls or complete
ban on the services. The OFT holds such controls as market disruptive (OFT, 2010).
Salleh et al. (2012) demonstrate that Islamic prohibition of interest in credit transactions
(ribā an-nasi‟ah) is attributed to the inclination for expropriation of wealth. This occurs if
there is inequity in the financial contract, thereby resulting in two equilibrium cases. One, in
the case of financial repression, where the real interest rate is negative, the lender‟s assets are
expropriated. Two, in case of negative leverage, where the real interest rate is greater than
the unleveraged expected return on the asset being financed, then the borrower‟s assets are
expropriated. In the long-run, this creates imbalances or unsustainable equilibria. When a
borrower defaults, this can create a domino effect, given the interconnectedness of credit
markets. It effectively amplifies volatility within the financial system and thus precipitates
financial fragility, as evidenced in the present crisis. In extreme cases of agency costs of debt
accruing to high project and default risks, this can lead to autarky or financial exclusion, with
adverse impact on the underprivileged. Their study demonstrates interest-based financial
transactions are non-Pareto optimal or at best it is Pareto-neutral to a risk-free credit.
Instead, the Qur‟ān (the Holy Book of Islam) contrasts interest (ribā) with that of
charity (sadaqah). The former creates factions in society, whilst the latter cements social
cohesiveness (Ibn Taymiyyah, 1951). Charity, as defined in the practice of Prophet
Muhammad PBUH (Sunnah), is not only concerned with financial forms but also all types of
good deeds (Sahih Bukhari Vol. 2, 24:144; Sahih Muslim Vol. 3, 12:2329 -2330).17
“Charities ...provided a source of support for institutions and interest groups independent of,
and sometimes in opposition, to the state. Islamic charities have historically played an
additional role in society, that of promoter of decentralized economic development. ...In this
respect, they reflect the blending of religious and secular, the social and economic, that is the
key characteristic of the Islamic idea” (Bremer, 2004, p.7). Piety through charitable deeds
inculcates a sense of brotherhood and advances social welfare. This is clearly demonstrated
in the Qur‟ān that censures the practice of creditors, who cumulate the amount due for every
delay in settlement that leads to further financial hardship on the debtor. The Qur‟ān calls for
the creditor to grant respite to the borrower such that, if the creditor were to forfeit the
amount owed, this reflects a higher order of virtuousness, and will be rightly rewarded (verse
2:280; Sunan Ibn Mâjah – translated al-Khattab, Vol. 3:2417-2420).
17
Although Islam enjoins charitable deeds, it prohibits begging, for it is best to be actively employed to uplift
one‟s economic status (Sahih Bukhari Vol. 2, 24:1470-1471; Sahih Muslim Vol. 3, 12:2396, 2400 and
2404).
Both the Qur‟ān and Sunnah have specific references for assisting the underprivileged.
The financial forms of charity can be broadly categorised into zakāt (social welfare funds),
waqf (philanthropic foundations) coupled with qard (interest-free loan) or salaf (synonymous
with interest-free loan). Zakāt forms one of the five pillars of Islam and is obligatory on
one‟s wealth for the benefit of the recipients identified in the Qur‟ān (verse 9:60). Of interest
is the specific directive for financial resources to be allotted for the poor and needy.
Although waqf (awqāf, plural) is not mentioned specifically in the Qur‟ān, it plays an
instrumental role in Islamic civilisation. The earliest records on the practice of waqf can be
traced to the Ottoman Empire in the eight century (Cizacka, 2000). It is said that these
philanthropic foundations were able to financially support the provision of social services in
Muslim society at that time, and in turn helped to address economic disparity. Such practice
involved the endowment of privately owned properties for charitable purposes in perpetuity.
The revenue generated by the waqf is then utilised according to its objects.
Qard signifies the extension of loan to a borrower from one‟s property without
expectation of gains, whereby the lender forfeits the use of his property during the loaned
period. Such is its prominence that it is ranked higher than charity and even equated as a loan
to God himself (verses 2:245; 5:12; 57:11 and 18; 64:17; 73:20).18
This benevolent loan is
also synonymous with salaf that connotes the extension of a loan, subject to repayment at a
later time (Al-Zuhayli, 2003). From a fiqh (Islamic jurisprudence) perspective, jurists are
divided on the rights of the lender on the terms of the loan. Two widely opposing views are
that a lender has absolute rights to recall the loan at anytime, whilst others view that it is
permissible for the lender to stipulate the loaned period and hence, both contracting parties
should abide by it (see Al-Zuhayli, 2003 on the debate by the four major Sunni schools of
thought). This ambiguity in the Shari‟ah interpretation can cause adverse repercussions in
current financial context that warrants property rights certainty.19
Underdevelopment of the
fiqh accompanied with inadequacies of the zakāt system is a barrier to realisation of its true
potential (see Kuran, 2003; 2004; 2011). This warrants reinvigorating the zakāt
implementing institution and instigating the development of fiqh, as discussed below.
18
Ali (2002) connotes qard to “spending in the cause of God” (footnote 710, p.245). A benevolent loan does
not exempt the borrower from honouring the debt. The severity of non-repayment is highlighted in the
Sunnah whereby even a martyr who is forgiven for every sin is still bound by his debt (Sahih Muslim Vol.
5, 33:4883-4884. See also Ibn Mâjah Vol. 3:2412-2413, 2431).
19 Earliest record on employment of qard by Az-Zubair also does not allude to its form and activity in which
it was deployed (Sahih Bukhari Vol. 4, 57:3129).
IV. Model development
This section details the mathematical design of an efficient interest-free short term
payday loan facility (using endogenous leverage) to address the inter-temporal liquidity needs
of unbanked and underbanked households. Our endogenous leveraged circuit is founded in
the works of institutional economics (Commons, 1931), and builds from the technology of
ROSCA (Besley et al., 1993; 1994; Gugerty, 2007), its associated hybrids; namely, ASCRA
(Bouman, 1995), and the more contemporary mutual and financial cooperative (Emmons and
Mueller, 1997; Ebrahim, 2009). Besides liquidity transformation, the circuit features akin to
an Islamic insurance (takāful) or mutual scheme where members guarantee each other from
unexpected damage, losses or misfortune (Bouman, 1995).
Furthermore, unlike other endogenous leverage groupings, liquidity constrained
members of the circuit receive short term interest-free payday loans, which is repaid at their
next payday date. Our model paper expounds the elements that need to be observed if an
interest-free loan that is enjoined in Islam is to have a profound impact in any financial
development schemes.20
Here, we demonstrate that this endogenous interest-free payday
loan circuit integrated with appropriate constraints that circumvent adverse selection and
moral hazard, can be Pareto efficient or at least neutral to that of its competitor, i.e.,
mainstream financier.
The interest-free payday loan circuit is structured as follows. Individuals are required
to become members by contributing monthly to a common pool of funds. In our model,
members are risk neutral, and the demand for liquidity or payday loan is treated as
exogenous. Members can only apply for the interest-free loan after qualifying a defined
period of membership. This gestation period has a two-fold effect. First, it allows the circuit
to identify and assist the member in realisation of her/his financial goals. Second, it allows
build up of member equity cushion through the monthly contributions. This effectively binds
the member to the circuit and addresses member time-inconsistent preferences. In addition to
these two covenants, other mandatory rulings to address adverse selection and moral hazard
issues (ie., default cost) include requirement for (i) direct deposit of member paycheque into
20
Our model explores alternative platforms for deployment of interest-free financing and augments previous
studies by (i) Darrat and Ebrahim (1999) who focus on open market operation instrument in a partial
equilibrium framework of qard-based Malaysian Government Investment Certificates; and (ii) Ebrahim
(2009) on structuring alternative solutions for long-term real estate financing.
the circuit; and (ii) existence of loan guarantor (see detailed explanation below).
Furthermore, once a member borrows from the circuit, she/he is required to undergo financial
planning program to enhance her/his financial literacy. This helps errant members to plan
ahead, alleviate future liquidity crises and stay debt free.
We implicitly assume the existence of an information architecture, where property
rights needed for the forthcoming paycheque to serve as collateral, accurate methods of
verifying or evaluating members‟ income and bankruptcy procedures are well established
(see Levine et al., 2000). Individuals joining the circuit, assumed to have limited asset
qualifying collateral and void of other alternative credit solutions, would have to subscribe to
current high cost payday loan. Each member receives an exogenous flow of income. These
assumptions are representative of the stylised facts of payday loan borrowers‟ demographics
(see Ellihausen and Lawrence, 2001; FDIC, 2009).
The institutional framework of our efficient interest-free endogenous leverage circuit
(i.e., Model 1), including essential covenants, is discussed in Section IVa. This is followed
by stepped extensions of this basic framework: (i) Model 2 in Section IVb incorporates
elements of (non)borrowing members within a circuit with a definite terminal period; and (ii)
Model 3 in Section IVc further relaxes the elements whereby there is random borrowing that
in the long-run approaches a steady state. A stylised depiction of the various scenarios for
each of the model is captured in Figure 1.
[Insert Figure 1 here]
IVa. Institutional framework of an interest-free short term financing circuit (Model 1 –
Figure 2)
[Insert Figure 2 here]
IVa(1). This framework embeds elements of ROSCA and ASCRA technology (see Besley et
al., 1993; 1994; Bouman, 1995; and Gugerty, 2007). The basic structure in Model 1
below illustrates the case of self-insurance, where circuit members are savers and
also liquidity constrained borrowers.
(i) Seed funding for the circuit stems from charitable funds (e.g. zakāt and sadaqah funds)
that assist in the circuit start up. Ideally for long-run stability, this charitable fund
should be institutionalised and performs the central role of providing liquidity relief to
individual circuits that may suffer from unforeseen shocks. This is akin to the Verband,
associative level of the German cooperative banking system (Biasin, 2010).
(ii) On joining the circuit, each member is required to make periodic monthly contributions
„C‟ into a pooled fund, from time i = -m (at the point of membership) to i = 1 (the
circuit terminal date). Here, we adopt monthly contributions to maintain consistency
with the members‟ income stream, i.e., paycheques are generally issued on monthly
basis. By instituting periodic contribution, it (i) moderates information asymmetry and
adverse selection, as it reveals the financial behaviour of the prospective borrower
(Akerlof, 1970); (ii) assists in long-run accumulation of wealth that minimises
exposures to exogenous income shocks; (iii) acts as an equity buffer that minimises the
likelihood of the member to strategically default on her/his borrowing (Foote et al.,
2008); and (iv) most importantly, it acts a commitment device that moderates self-
control issues associated with payday loan borrowers (Gugerty, 2007; Skiba and
Tobacman, 2008).
The accumulated periodic contribution, represented by „S‟, forms the capital base of
this circuit and is used to meet short term financial needs of liquidity constrained
members. This is given in Equation (1), where „m‟ denotes the month building up to
the disbursement of the interest-free liquidity facility.
t
mi
mtCCS )1(
(1)
When 1t (as in Models 1 and 2), this culminates into
1
)2(mi
mCCS (1a)
(iii) After satisfying the minimum gestation period, liquidity constrained members qualify
to draw „Q‟ interest-free short term financing from the circuit at time i = 1-n, where „n‟
is a fraction of a month (i.e., n < 1 month).21
The interest-free facility resembles a
bullet loan, where total repayment of principal „Q‟ is made at terminal time i = 1. The
loan repayment is net off the accumulated contributions „S‟.
SQ (2)
21
This implies members face liquidity problems before their next paycheque.
(iv) To account for the opportunity cost of capital employed within the circuit, „C‟ and „Q‟
are discounted by „γ‟. That is, the monthly discount rate comprises of an imputed
return „r‟, which is equivalent to the average cost of fund incurred in mainstream credit
market.
r
1
1 < 1, r > 0 (3)
(v) We also incorporate transaction cost „ζ‟ associated with administering the circuit, eg.
managing of members‟ contributions and loan processing (Emmons and Mueller, 1997;
Kontolaimou and Tsekouras, 2010), and default cost „α‟ (Jaffee and Russell, 1976).22, 23
The circuit efficiency is contingent on minimising transaction and default costs, as they
can fritter away the circuit‟s gains or cause erosion to its capital base (Coase, 1937;
Alchian, 1950; Hart and Moore, 1996). In the models, both outflows are moderated by
presence of covenants discussed below.
In line with the circuit‟s objectives, the discounted contributions and interest-free loan
after accounting for transaction and default costs, coupled with net loan payoff, given
by Equation (4) should at least be equal or greater than zero.
0111 11
SSQQC ni
mi
(4)
Substituting the value of „S‟ in Equation (1) into Equation (4) gives us:
012211...11 11 mCQQC nmm (4a)
The periodic contribution in Equation (4a) form a geometric series that can be further
simplified as follows:
22
The institutional structure of the circuit already minimises upfront transaction costs compared to current
payday lenders, as it (i) benefits from non-profit motive management force, (ii) does not incur external
funding costs, and (iii) is not bound to issue investment returns to its „depositors‟.
23 Intuitively, utility derived from an interest-free credit facility would be higher than subscribing to high cost
current payday loans or face financial autarky from mainstream credit. Therefore in such situations, we
foresee that the (non) pecuniary costs associated with default penalty should be significantly severe such
that it impels repayment of the loan (see Skiba and Tobacman, 2008 for empirical evidence of the degree
of reliance of these borrowers on payday loans for their liquidity needs).
111221
11 12
nmm
QmC
(4b)
11
1122
1
111
2
n
mm
mCQ
(4c)
Potential maximum loan is,
11
1122
1
111
2
1max n
mm
mCQ
(4d)
IVa(2). To moderate the risk of adverse selection and moral hazard, it is imperative for the
circuit to institute covenants as follows:
(i) Income and loan constraint: Each member is subjected to an after tax income test „y‟
to ascertain her/his capacity to meet her/his periodic contribution and loan obligation.
This not only supports responsible lending (Carr and Kolluri, 2001; Morgan, 2007), but
also moderates the adverse selection issues (Jaffee and Russell, 1976; Stiglitz and
Weiss, 1981). The member‟s financial capacity is represented by a multiple „b‟ of
her/his income and loan.
(ia) Income constraint „b1‟
Here, the income constraint „b1‟ curtails the contribution „C‟, given as follows:
1bC
y , which can be rewritten
1b
yC (5)
1
maxb
yC (5a)
(ib) Loan constraint „b2‟24
Here, the loan constraint „b2‟ curtails the loan amount „Q‟, given as follows:
2bQ
y , which can be rewritten
2b
yQ (6)
Here,
ntntBinding Qb
yQ max,
2
, ,min , where ntnt QQQ ,2max1maxmax, ,min (6a)
1maxQ is defined in Models 1 – 3 (sections IVa – IVc) respectively by Equations (4d),
(7e) and (16d), while ntQ ,2max reflects the resource constraint of the circuit given by
mtCQ nt 1,2max .
24
The interest-free payday loan facility is strictly for managing inter-temporal liquidity shocks faced by its
members. In tandem with this objective, „Q‟ should therefore be confined to a reasonable multiple of its
members‟ monthly after tax income. This helps alleviate debt entrapment, discussed in Ellihausen and
Lawrence (2001). Nonetheless, our model can still be adapted to reflect allowances for this restriction.
If transaction cost is low, i.e., max
2
Qb
y , then
2
,b
yQ ntBinding (6b)
(ii) Pre-commitment constraint: Members are subjected to salary deduction to moderate
time inconsistent preference tendencies (Liabson, 1997; Stephens, 2002; Gugerty,
2007; Skiba and Tobacman, 2008) and moral hazard. With this, it partially limits the
member consumption options available in the future. This seamless transfer of member
income to the circuit and subsequent settlement of the interest-free loan has a secondary
effect of lowering transaction costs of the circuit.
(iii) Collateral constraint: Given the limited ability of unbanked and underbanked
households to raise asset qualifying collateral, disbursement of the interest-free payday
loan is then subjected to a reputable co-signer, who provides surety upon default by the
member.25
The co-signer, who has local information compared to the circuit, is in a
preferred position to conduct ex-post monitoring and impose social sanctions.26
This
then, significantly reduces costly state verification issues (Ghatak, 1999), particularly in
dealings with low net worth members. However, failure of the co-signer to act
accordingly can have a detrimental effect on the circuit efficiency/ sustainability (see
Guinnane, 1994 on demise of Irish credit union).
(iv) Financial capability constraint: Each member who borrows is required to undergo
personal finance program (eg. money management, asset building and debt
management) to enhance their financial capability (Agarwal et al., 2009; Gathergood,
2012). This non-pecuniary cost of borrowing is an interventionist measure that has its
roots in behavioural finance, as it seeks to influence the cognitive psychology of
payday borrowers with regards to their financial conduct (see Bernheim et al., 2001;
25
The permissibility of surety is affirmed in the Sunnah and by jurists (Ibn Rushd, 1996; Ibn Mâjah Vol. 3,
Chapter on guaranty, No.2405-2406).
26 The collateral covenant permits the co-signer to be from or outside the circuit. Where the co-signer is also
a member of the circuit, co-signing incentivises peer monitoring, in view that the sustainability of the
circuit ultimately affects the interest of the consignor (Stiglitz, 1990).
Bernheim and Garrett, 2003 on the positive long-term behavioural effects of increased
exposure to financial education).27
Proposition 1. For the circuit to be competitive, its net surplus must satisfy the efficiency
condition given by Equation (4c)
Equation (4c) signifies three possible states of the circuit. First, when the circuit fulfils
the equality sign, the circuit is at best Pareto-neutral to its competitors, namely mainstream
financiers.28
Second, if the inequality sign is satisfied, the circuit then is Pareto-superior to
its competitors. The surplus capital signifies welfare improvement of an initially liquidity
constrained group. Third, if Equation (4c) is unmet, then the circuit is Pareto-inferior with
erosion in its capital base, and its continued sustainability is doubtful. Here, its sustainability
is contingent on minimising transaction costs and defaults, as both erode the circuit‟s gains
and ultimately its capital base. Therefore, the circuit administrators must institute controls, so
that both costs are reduced significantly. This is achieved through various covenants, as
highlighted in the paper.
IVb. Extension of the model with fraction of members ‘λ’ borrowing from the circuit
(Model 2 – Figure 3)
[Insert Figure 3 here]
As in Model 1, members are required to contribute „C‟ on monthly basis upon entry, at
i = -m to i = 1 period. The following similarly hold in Model 2: (i) the circuit has a defined
period, i.e., until i = 1, after which it terminates; (ii) variables defined in Equations (1), (2),
(3) and (4); (iii) transaction and default costs; and (iv) the four covenants (i.e., income, pre-
commitment, collateral and financial capability). However, Model 2 specification differs
from previous in that it conceives that the likelihood of liquidity strained members may occur
at different circuit cycles. Therefore, at any one time, there are a proportion of borrowers
signified by „λ‟ that are supported by „(1-λ)‟ lenders. This clearly depicts the „transformation
service‟ provided by the circuit, whereby the temporary idle funds of a proportion of
27
Other interventionist measures, which are pecuniary in nature, are to gradually: (i) decrease b1, and (ii)
increase b2, thereby compelling erring borrowers to save and avoid debt entrapment. This can be extended
in our model to incorporate real world practicalities.
28 We can also deduce that the circuit is Pareto-superior to that of contemporary payday loan, in view of the
latter‟s high cost of funds.
members (lenders or non-borrowers) are used to provide financing to others who suffer from
exogenous inter-temporal income shocks. This improves on “competitive market by
providing better risk sharing among people who need to consume at different random times”
(Diamond and Dybvig, 1983, p.402).
As with the previous section, observance of „C‟ entitles a member to the right of
drawing on the circuit funds if she/he faces liquidity squeeze. We find the technology of the
circuit in Model 2 best resembles the practice of group insurance or Islamic insurance
(takāful), where members agree to indemnify each other against a defined loss. Based on the
concept of solidarity, members of the group contribute to a specified fund that entitles each
person to protection on occurrence of the loss event. The commercial implementation of this
concept of mutuality can be traced to the eight century, where sea merchants would initiate a
pool to protect themselves against perils during their voyages (Alhabshi and Razak, 2011).29
A characteristic that differentiates group/ mutual/ Islamic insurance from the mainstream is
that, in the former, each member is the insurer and also insured, which means there is risk
sharing between members rather than risk shifting.
Based on the above extension, total borrowings in the circuit are now signified by „λQ‟.
At time i = 1, i.e., expiry of the circuit cycle: (i) borrowing members are required to settle the
outstanding interest-free short term loan net of accumulated contributions „λ(Q-S)‟; and (ii)
proportion of non-borrowing members „(1-λ)‟ are then entitled to a payback of their
accumulated contributions constituting „(1-λ)S‟.30
Equation (7) is a modification of Equation
(4), as it incorporates the fraction of borrowing and non-borrowing members.
01)(111 11
SSSQQC ni
mi
(7)
0)12(11 11
SQQQC ni
mi
(7a)
29
Although there is no direct reference to takāful in Islamic scriptures, the concept finds support in the
Qur‟ānic verses and Sunnah that call for the removing of hardships, upholding of brotherhood and
solidarity in the face of hardship (Qur‟ān 5:2; Sahih Muslim Vol. 6, 45: 6585-6590, 45:6669-6674).
30 In our model, a member‟s primary objective in joining the circuit is to ensure access to credit, in contrast to
excessive credit pricing by payday lenders or financial autarky. Thus, we assume the only incentive is to
maximise borrowing rather than generate dividends or returns on the accumulated contributions.
Substituting the value of „S‟ in Equation (1) into Equation (7a) gives us:
012211......11 11 mCQQC nmm
(7b)
This is further simplified as follows:
1)1(1221
11 12
nmm
QmC
(7c)
11
1122
1
111
2
n
mm
mCQ
(7d)
Potential maximum loan is,
11
1122
1
111
2
1max n
mm
mCQ
(7e)
Proposition 2. The efficiency condition of the circuit with fraction of members who are
(non) borrowers is contingent on satisfaction of Equation (7d)
The three efficiency states described earlier in Proposition 1 apply in Proposition 2.
This is even with the added complexity, where not all members will borrow during the same
circuit cycle. The circuit can ensure that it satisfies the inequality sign in Equation (7d) by
enhancing its predictive ability on probability of liquidity calls by its members. This is
closely associated with the principle law of large numbers employed in insurance pricing. By
collating sufficiently large number of exposures, the randomness in the occurrence of the
exposures will statistically converge towards a defined mean with a given variance, which
then allows insurers to fairly predict the frequency and severity of their exposures and price
the insurance products accordingly. In the case of the circuit, it can then correctly determine
the loan amount and tenure that is feasible to limit liquidity gaps at the end of the circuit
cycle.
IVc. Extended model incorporating random multi-period borrowing (Model 3 – Figure
4)
[Insert Figure 4 here]
We further extend the model to allow for multi-period endogenous leverage, where
members pool their endowments across time, in order to assure accessibility to short-term
interest-free loan, in light of unexpected contingencies. This brings the circuit nearer to that
of contemporary financial cooperatives (Emmons and Mueller, 1997; Ebrahim, 2009). Here,
we have a random process of member borrowing. This discrete-parameter Markov chain of
{Xt-n, t>n} is represented by:
)/(),....,,/( 122111 tntnttntnnnt iXjXPiXiXiXjXP (8)
Equation (8) essentially assumes a member‟s future borrowing behaviour is a
consideration of only her/his present behaviour, and is independent of the member‟s past
history. The initial probability vector „pt-n‟ is denoted by probability of borrowing „λt-n‟ and
non-borrowing „1-λt-n‟, respectively:
ntntntp 1 (9)
We also assume during the next interval that there is a probability „ρt+1-n‟ that members
borrow and „1-ρt+1-n‟ otherwise. The two-state Markov chain transition probability matrix is
illustrated below.
Borrowt+1-n NoBorrowt+1-n
nt
nt
NoBorrow
Borrow
P
ntnt
ntnt
11
11
1
1
(10)
The above two-state transition matrix converges in steady state as follows (see Hsu,
2011).
1
1
P (10a)
This matrix in Equation (11a) is further simplified using the well-known Bayes‟ rule, as
illustrated in the Appendix, where we realise
1
1, and
1
211 .
Borrowt+1-n NoBorrowt+1-n
nt
nt
NoBorrow
Borrow
P
1
21
1
1
1
(10b)
The long-run borrowing behaviour converges to a steady state „ ̂ ‟. That is, there exists
a stationary distribution for the Markov chain. This is found by solving
pp ˆ
1
21
1
1
1
ˆ
(11)
Where, 21ˆ ssp , and 121 ss (12)
Equation (11) can then be rewritten as follows:
21 ss
21
1
21
1
1
1
ss
(13)
Solving the matrix, we obtain two equations described below:
Equation 1
121
1
1sss
(14)
121 ss , for ρ ≠ 1 (14a)
Equation 2
2211
21)1( sss
(15)
121 ss , ρ ≠ 1 (15a)
Thus, both Equations (14a) and (15a) lead to the same solution, implying the
exogeneity of „ρ‟. By substituting „s1‟ in Equation (14a) from Equation (12), we get
11 22 ss
(15c)
1
11 2s
12s , and hence 1s (15d)
Thus, restating „ ̂ ‟ of Equation (11) with the results derived in Equation (15d) gives us
the steady state matrix as follows:
1ˆ (15e)
Proposition 3. A member borrowing behaviour is contingent on her/his past borrowing
history.
We find member borrowing behaviour is path dependent, which corroborates the
empirical evidence documented in Elliehausen and Lawrence (2001). Despite this intricate
issue of path dependency, we can still determine the loans to be underwritten by exploiting
the property of steady state, where a fraction „λ‟ of the population borrow (irrespective of
previous borrowing). For mathematical tractability, we assume that: (i) borrowers who do
not redeem their loans, continuously rollover their facility; and (ii) a fraction „α‟ of the loans
„Q‟ is written off as default at the terminal period T (see Figure 1). This is aligned with
Elliehausen and Lawrence (2001), whereby a fraction of the borrowing pool exhibits
tendency of rolling over their loans. As such, default will emerge only at the termination of
the program. This is illustrated in Equation (16) (given below). All other variables and
covenants remain the same.
0)12()1(111
1
SQQQC TT
nti
inT
nti
iT
mi
i
(16)
We substitute the value of „ S ‟ from Equation (1) into Equation (16) and simplifying it
to derive
ntTnttTntmTm QQC 21 ...1...1)(1...1)(1
0)12)(1()1( mTCQT (16a)
1
1
1
1)(1
1
1)(1
11
11 ntTnt
tTnt
mTm QQC
0)12)(1()1( mTCQT (16b)
1211
11 1
mTCQ T
mTm
X
1
1
111
1111 TntTnttTnt
(16c)
Potential maximum loan is,
1211
11 1
1max
mTCQ T
mTm
X
1
1
111
1111 TntTnttTnt
(16d)
Proposition 4. The efficiency condition of a circuit with multi-period borrowing by a
fraction of members is contingent on satisfaction of Equation (16c)
The efficiency states detailed in Proposition 1 similarly applies for Proposition 4. We
find that a circuit‟s efficiency can be improved in a multi-period model. A circuit that is
conducted repeatedly over a series of periods will have greater latitude on its borrowing
policy, as each borrower‟s financial conduct is fully revealed (Hosios and Peters, 1989). By
instituting renewal model that is dependent on the member‟s financial conduct, the circuit
effectively addresses conflict of interest between borrowers and non-borrowers. In this
situation, each member will endeavour to undertake fewer risks, which would potentially
affect access to future liquidity facility. Credible threat of sanctions in multi-period states
can also reduce moral hazard (Stiglitz and Weiss, 1983).
IVc(1). Numerical illustration
Using Equations (4d), (7e) and (16d), we conduct a mathematical simulation to
enumerate the breakeven level of the loan advanced in each of the three models. The circuit
exogenous factors encompass: (i) member income profile y; (ii) cost of fund prevailing in
mainstream credit market r; (iii) transaction and default costs ζ and α; (iv) loan tenure and
drawdown period n and t; (v) underwriting constraint corresponding to the income multiple
b1 and b2; (vi) gestation period prior to loan drawdown m; (vii) fraction of borrowing
members λ; and (viii) substantive circuit life T. We use the observations by Ellihausen and
Lawrence (2001), the FDIC (2010) and Hawkins (2011) to check the reasonableness of the
exogenous parameters. Overall, the final values of the exogenous parameters are set to avoid
excessive financial burden and ensure a liberal round of liquidity cycle, until member reaches
financial security.
[Insert Table 1 here]
We tabulate the efficiency scenarios, given various permutations of the endogenous
parameters, which cover (i) maximum member monthly contribution: Cmax; (ii) maximum
accumulated savings: Smax; and (iii) potential maximum loan: Qmax. Table 1 illustrates the
effect on the endogenous factors, given changes in the exogenous parameters. This provides
an indicative pricing framework that can be emulated in the design of similar endogenous
circuits. It highlights the sensitivity of each endogenous factor to the decisions that the
circuit undertakes and the various levers that may be combined to enhance the circuit
efficiency.
Additionally, Table 2 provides the resultant values of the endogenous parameters,
which assure that the circuit satisfies the Pareto-efficiency propositions under Models 1 - 3.
Model 3 further demonstrates the interplay of the loan constraint covenant between 2b
y or
Qmax, where Qmax is characterised by the lower of either Qmax1 or Qmax2, t-n. Here, QBinding
demands balancing the twin issues of: (i) protecting the member from potential debt
entrapment; and (ii) ensuring the circuit‟s long-run liquidity, i.e. solvency. By not pursuing
aggressive loan disbursement policies, it promotes accumulation of equity buffer that would
ultimately allow the circuit greater financial latitude to pursue financial policies that enhance
member welfare within the reasonable risk tolerance limits, eg. relaxing the „QBinding‟
constraint and undertaking loan rehabilitation program that customises the loan repayment
tenure for genuinely financially constrained member.
[Insert Table 2 here]
Based on Equation (16d), we extend the simulation to illustrate the effect of transaction
cost on the potential loan amount, while holding other exogenous factors constant (see Figure
4).31
Premised on the loan covenant in Equation (6) with loan multiple of 102 b and
monthly after tax income of 000,2$y , 2b
yis then fixed at $200. On the other hand, „Qmax‟
changes with variation in the transaction cost „ζ‟, i.e., there exists an inverse relationship
between „Qmax‟ and „ζ‟. „ BindingQ ‟ as given in Equation (6a) is the minimum of either2b
y or
Qmax. As highlighted in Equation (6b), if the transaction cost is low, i.e., MaxQb
y
2
, then
2b
yQBinding . Otherwise, QBinding is restrained by Qmax. The critical transaction load, whereby
2b
y = Qmax, is when %7.28 , which is signified in Figure 4 by critical .
[Insert Figure 5 here]
V. Discussion and conclusion
Payday borrowers are categorically those who suffer from poor credit history, exhibit
time-inconsistent preferences and are often precluded by mainstream financiers. This
financial exclusion and poor financial tendencies expose them to potential predation by
payday lenders who offer short term, small and unsecured financing at usurious rates. Given
payday borrowers‟ already precarious financial standing, the repressive terms (i.e., high cost
lending and predatory motives) could lead to a vicious cycle of debt entrapment if the
borrower fails to observe the repayment term. Despite the unfavourable publicity against
payday loans industry, financially constrained households still succumb to its services. This
31
Exogenous factors are: y = $2,000, m = 6 months, r = 15%, α = 10%, λ = 0.4, T = 180 months, and t = 1.
underlines a latent need for inexpensive short term financial facility to bridge their liquidity
needs.
Unlike previous studies on payday loans, this study undertakes to conceptualise a
solution to usurious payday loans and address financial autarky in mainstream credit. It
entails the design of an institutional structure that embeds the interplay of cultural beliefs and
cost efficient organisations. Our model is based on risk-neutral economic agents within an
endogenous leverage circuit that draws from the technology of member based institutions
such as ROSCA and its hybrids, i.e., ASCRA, mutual and financial cooperative. A unique
feature of this structure is that it harnesses the concept of coalition of savers and borrowers to
allay inter-temporal liquidity shocks faced by its members, through the deployment of
interest-free payday loan.
Our study illustrates the employment of this antiquated charitable form in contemporary
financial perspective. This credit modality is chosen for its contrast with current payday
loans. Furthermore, it is held esteemed in religious tenets, which specifically distinguish
such deeds from usurious practices. We demonstrate that the circuit performs favourably in
contrast to current usurious payday loans. First, the interest-free facility averts expropriation
of wealth, an issue associated with payday loans. The periodic contribution, which features a
minimal fraction of members‟ salary, promotes asset building which should consequently
improve their financial security in the long-run. Second, our member based endogenous
circuit attenuates the issue of financial exclusion. It allows credit accessibility to households
disfavoured by financial institutions. This financial reach extends not only to those
underbanked but even unbanked households that would typically be excluded from
mainstream credit markets. Third, we attest the economic proposition of interest-free loans
expounded in religious teachings in current financial settings, whereby the circuit is able to
boot strap its resources to grow endogenously. Fourth, in line with documented studies of
time-inconsistent preferences of these households, the institutional design of our interest-free
payday loan relies on the commitment technologies advocated with circuit-based structures.
The efficiency of an endogenous leveraged circuit is contingent on observing risk
control measures to constraint adverse selection and moral hazard issues, thus reducing
default and transaction cost significantly. This builds on ensuring equitable commitment, i.e.,
the periodic contribution can be fulfilled without jeopardising financial interests of individual
members and the circuit. This is followed through by requiring the commitment to be
directly dispensed into the circuit, which would effectively pre-empt irrational consumption
tendencies. This is fortified with programs that build the members‟ financial capability and
route them from poor credit tendencies. Next, we require existence of co-signer that acts in
absence of standard collateral. Here, the co-signer‟s central role is in reducing costly state
verification and execution of credible sanctions. The institutional design of the circuit
provides upfront dilution of transaction costs that directly feeds into promoting the circuit
efficiency.
Given the circuit technology that is member driven, accumulation of substantive capital
base may create lag in time to loan issuance. To manage the gestation period, the circuit may
rely on seed funding from zakāt and sadaqah to reinforce its initial capital base. These
charitable funds can also be institutionalised to provide safety net to the circuit that defrays
any long-run sustainability issues. Alternatively, the interest-free loan facility can be
integrated into an already operational circuit. Results from our study supports the policy
direction outlined in the FDIC (2007) small-dollar loan program. That is, it adopts affordable
pricing, observes risk-based underwriting, maximises technology and automation, integrates
savings component in combination with financial education. Last, our model sets an
indicative pricing mechanism, mostly absent in charitable institutions, which in the long-run
promotes self-sufficiency.
References
Al-Bukhārī, Abū „Abdullāh Muhammad bin Ismā‟īl bin Ibrāhīm bin Al-Mughīra (1997). The
translation of the meanings of Sahih al-Bukhārī (translated by Khan, M. M.). Riyadh,
Saudi Arabia: Darrusalam.
Al-Zuhayli, W. (2003). Financial transactions in Islamic jurisprudence, Vol. 1. Damascus,
Syria: Dar al-Fikr.
al Alwani, T. J. (1993). Ijtihad (1st ed.). Herndon, Virginia: International Institute of Islamic
Thought.
Alchian, A.A. (1950). Uncertainty, evolution and economic theory. Journal of Political
Economy 58, 211-221.
Alhabshi, S.O. and Razak, S.H. (2011). Takaful insurance: Concept, history and development
challenges. In Ariff, M. and Iqbal, M. (Ed.), The foundations of Islamic banking: Theory,
practice and education. Cheltenham, UK: Edward Elgar Publishing, 184-207.
Akerlof, G. A. (1970). The market for “lemons”: Quality uncertainty and the market
mechanism. The Quarterly Journal of Economics 84, 488-500.
Ali, A. Y. (2002). The Holy Qur‟an: Text, translation and commentary. Elmhurst, New York:
Tahrike Tarsile Qur‟an Inc.
Agarwal, S., Skiba, P.M., and Tobacman, J. (2009). Payday loans and credit card: New
liquidity and credit scoring puzzles? NBER Working Paper Series 14659, (January).
Baltensperger, E. (1978). Credit rationing: issues and questions. Journal of Money, Credit
and Banking 10, 170-183.
Beck, T. and Demirguc-Kunt, A. (2007). AccessFinance. A Newsletter Published by the
Financial and Private Sector Development Vice Presidency, (June).
Bertrand, M., and Morse, A. (2011). Information disclosure, cognitive biases, and payday
borrowing. The Journal of Finance, 66, 1865-1893.
Bernheim, B.D., Garrett, D.M., and Maki, D.M. (2001). Education and saving: the long-term
effects of high school financial curriculum mandates. Journal of Public Economics 80,
435-465.
Bernheim, B.D., and Garrett, D.M. (2003). The effects of financial education in the
workplace: evidence from a survey of households. Journal of Public Economics 87,
1487-1519.
Besley, T., Coate, S., and Loury, G. (1993). The economics of rotating savings and credit
associations. The American Economic Review 83, 792-810.
Besley, T., Coate, S., and Loury, G. (1994). Rotating savings and credit associations, credit
markets and efficiency. The Review of Economic Studies 61, 701-719.
Biasin, M. (2010). The German cooperative banking system. In V. Boscia, A. Carretta, and
P. Schwizer, Cooperative banking in Europe: Case studies. New York: Palgrave
Macmillan, 129-147.
Bouman, F.J.A. (1995). Rotating and accumulating savings and credit associations: A
development perspective. World Development 23, 371-384.
Bremer, J. (2004). Islamic philanthropy: Reviving traditional forms for building social
justice. Paper presented at the CSID Fifth Annual Conference on “Defining and
Establishing Justice in Muslim Societies”, Washington, D.C., USA.
Burton, M. (2010). Keeping the plates spinning: Perceptions of payday loans in Great
Britain. UK: Consumer Focus, (August).
Carr, J.H., and Kolluri, L. (2001). Predatory lending: An overview. Washington, D.C., USA:
Fannie Mae Foundation.
Campbell, J.Y., Jackson, H.E., Madrian, B.C., and Tufano, P. (2010). The regulation of
consumer financial products: An introductory essay with four case studies. Harvard
Kennedy School Faculty Research Working Paper Series, RWP10-040, (September).
Caskey, J.P. (October 2010). Payday lending: New research and the big question. Federal
Reserve Bank of Philadelphia Working Papers 10-32. Philadelphia: Federal Reserve
Bank of Philadelphia.
Caskey, J.P. (2005). Fringe banking and the rise of payday lending. In P. Bolton, and H.
Rosenthal (Eds.), Credit markets for the poor (p. 17-45). USA: Russell Sage Foundation.
CFSA. (2011). Community financial services association. Retrieved December, 2011, from
www.cfsaa.com
Cizakca, M. (2000). A history of philanthropic foundations: The Islamic world from the
seventh century to the present (1st ed.). Turkey: Bogazici University Press.
Coase, R.H. (1937). The nature of the firm. Economica 4, 386-405.
Commons, J.R. (1931). Institutional economics. American Economic Review 21, 648-657.
Darrat, A.F., and Ebrahim, M.S. (1999). On the design of interest-free instruments. Journal of
King Abdul Aziz University: Islamic Economics 8, 51-60.
DeYoung, R., and Phillips, R.J. (2009). Payday loan pricing No. 09-07. Kansas City: Federal
Reserve Bank of Kansas City, (February).
Diamond, D.W. and Dybvig, P,H. (1983). Bank runs, deposit insurance, and liquidity.
Journal of Political Economy 91, 401-419.
Ebrahim, M.S. (2009). Can an Islamic model of housing finance cooperative elevate the
economic status of the underprivileged? Journal of Economic Behavior and
Organization 72, 864-883.
El-Gamal, M.A. (2009). Islamic finance: Law, economics, and practice. New York:
Cambridge University Press.
Elliehausen, G., and Lawrence, E.C. (2001). Payday advance credit in America: An analysis
of customer demand (Monograph no. 35 ed.). Washington, D.C.: Credit Research Center
McDonough School of Business, (April).
Emmons W.R., and Mueller, W. (1997). Conflict of interest between borrowers and lenders
in credit cooperatives: The case of German co-operative banks. St. Louis, USA: Federal
Reserve Bank of St. Louis.
FDIC. (2007). Press release: Affordable small dollar loan guidelines.
http://www.fdic.gov/news/news/press/2007/pr07052a.html
FDIC. (2009). National survey of unbanked and underbanked households. USA: Federal
Deposit Insurance Corporation, (December).
FDIC. (2010). A template for success: The FDIC‟s small-dollar loan pilot program. FDIC
Quarterly, Vol. 4, No.2.
Flannery, M., and Samolyk, K. (2005). Payday lending: Do the costs justify the price? FDIC
Center for Financial Research Working Paper 2005-2009. USA: Federal Deposit
Insurance Corporation, (June).
Foote, C. L., Gerardi, K., and Willen, P. S. (2008). Negative equity and foreclosure: Theory
and evidence. Journal of Urban Economics 64, 234-245.
Gathergood, J. (2012). Self-control, financial literacy and consumer over-indebtedness.
Journal of Economic Psychology 33, 590-602.
Ghatak, M. (1999). Group lending, local information and peer selection. Journal of
Development Economics 60, 27-50.
Gugerty, M. K. (2007). You can‟t save alone: Commitment in rotating savings and credit
associations in Kenya. Economic Development and Cultural Change 55, 251-282.
Guinanne T.W. (1994). A failed institutional transplant: Raiffeisen‟s credit cooperatives in
Ireland, 1894-1914. Explorations in Economic History 31, 38-61.
Goff, S., and O'Connor, S. (2011), Payday lenders cross the Atlantic to target cash-strapped
Britons. Financial Times (December 7), 15.
Hawkins, J. (2011). Regulating on the fringe: Reexamining the link between fringe banking
and financial distress. Indiana Law Journal 86, 1361-1408.
HM Treasury (2012). Financial services > Financial services policy agenda > Mutuals.
Retrieved 13 July 2012, from the United Kingdom HM Treasury website
http://www.hm-treasury.gov.uk/fin_sector_mutuals_index.htm
Hosios, A. J., and Peters, M. (1989). Repeated insurance contracts with adverse selection and
limited commitment. Quarterly Journal of Economics 104, 229-253.
Hsu, H. (2011). Probability, random variables, and random processes. USA: Schaum‟s
outline series.
Ibn Majah Al-Qazwinî, Imâm Muhammad bin Yazeed (2007). Sunan Ibn Mâjah Vol. 3,
(translated by al-Khattab N. and edited by Zubair „Ali Za‟I, H.A.T.). Riyadh, Saudi
Arabia: Darussalam.
Ibn Rushd. (1997). Bidayat al-mujtahid wa-nihayat al-muqtasid: The distinguished jurist‟s
primer Vol. II, (translated by Imran Ahsan Khan Nyazee and edited by Mohammad
Abdul Rauf). Lebanon: Garnet Publishing Ltd.
Ibn Taymiyah, Taqi al-Din Ahmed bin „Abd al-Halim. (1951). Al-Qawā‟id al-Nū rā niyya al-
fiqhiyya. In M. al-Fiqi (Ed.). Cairo, Egypt: Al-sunna al-muhammadiyya, 112-113.
Jaffee, D.M., and Russell, T. (1976). Imperfect information, uncertainty, and credit rationing.
Quarterly Journal of Economics 90, 651-666.
Johnson, R.W., and Johnson, D.P. (1998). Pawnbroking in the U.S.: A profile of customers
(34th ed.). Washington: Credit Research Center.
Kontolaimou, A. and Tsekouras, K. (2010). Are cooperatives the weakest link in European
banking? A non-parametric metafrontier approach. Journal of Banking and Finance 34,
1946-1957.
Kuran, T. (2003). Islamic redistribution through zakat: Historical record and modern realities.
In M. Bonner, M. Ener and A. Singer, Poverty and charity in Middle Eastern contexts.
Albany, New York: SUNY Press, 275-293.
Kuran, T. (2004). Why the Middle East is economically underdeveloped: Historical
mechanisms of institutional stagnation. The Journal of Economic Perspectives 18, 71-90.
Kuran, T. (2011). The long divergence: How Islamic law held back the Middle East.
Princeton, Oxford: Princeton University Press.
Lawler, G.F. (1995). Introduction to stochastic processes. USA: Chapman & Hall Probability
Series.
Levine, R., Loayza, N., and Beck, T. (2000). Financial intermediation and growth: Causality
and causes. Journal of Monetary Economics 46, 31-77.
Lewinson, M. (1999). Conflicts of interest? The ethics of usury. Journal of Business Ethics
22, 327-339.
Liabson, D. (1997). Golden eggs and hyperbolic discounting. Quarterly Journal of
Economics 62, 443-477.
Mehr, R.I. and Cammack, E. (1972). Principles of Insurance. Homewood, Illinois: Richard D.
Irwin, Inc.
Melzer, B.T. (2011). The real costs of credit access: Evidence from the payday lending
market. Quarterly Journal of Economics 126, 517-555.
Melzer, B.T., and Morgan, D.P. (2009). Competition and adverse selection in the small-
dollar loan market: overdraft versus payday credit No. 391. New York: Federal Reserve
Bank of New York Staff Reports.
Moore, E. (2012). Credit unions benefit from rule changes. Financial Times (January 10), 4.
Morgan, D.P., Strain, M.R., and Seblani, I. (2012). How payday credit access affects
overdrafts and other outcomes. Journal of Money, Credit and Banking 44,519-531.
Morgan, D. P. (2007). Defining and detecting predatory lending (Staff Reports No. 273).
New York: Federal Reserve Board of New York.
Morse, A. (2011). Payday lenders: Heroes or villains? Journal of Financial Economics 102,
28-44.
Muslim, Imam Abul Husain bin Hajjāj (2007). English translation of Sahih Muslim
(translated by Nasiruddin al-Khattab, edited by Hâfiz Abu Tâhir Zubair „Ali Za‟i and
Huda Khattab). Riyadh, Saudi Arabia: Darussalam.
OFT. (2010). Review of high-cost credit: Final report. UK: Office of Fair Trading, (June).
OUP. (2012). “Loan” Dictionary of Finance and Banking in Economics and Business.
Oxford University Press Online.
Rahman, F. (1964). Riba and interest. Karachi: Islamic Studies.
Salleh, M.O., Jaafar, A., Ebrahim, M.S. and Molyneux, P. (2012). The prohibition of interest
(riba an-nasi‟ah) and the economic underdevelopment of the Muslim world. Working
paper, Bangor Business School, North Wales, UK.
Skiba, P.M., and Tobacman, J.B. (2009). Do payday loans cause bankruptcy? Vanderbilt Law
and Economics Research Paper No. 11-13, (November).
Skiba, P.M., and Tobacman, J.B. (2008). Payday loans, uncertainty and discounting:
Explaining patterns of borrowing, repayment, and default. Vanderbilt Law and
Economics Research Paper 08-33.
Skiba, P.M., and Tobacman, J.B. (2007). The profitability of payday loans. Unpublished
manuscript, (December).
Smith, T.E., Smith, M.M., and Wackes, J. (2008). Alternative financial service providers and
the spatial void hypothesis. Regional Science and Urban Economics 38, 205-227.
Stegman, M.A. (2007). Payday lending. The Journal of Economic Perspectives 21, 169-190.
Stiglitz, J.E. and Weiss, A. (1981). Credit rationing in markets with imperfect information.
The American Economic Review 71, 393-410.
Stiglitz, J.E. (1990). Peer monitoring and credit markets. The World Bank Economic Review
4, 351-366.
Zinman, J. (2010). Restricting consumer credit access: Household survey evidence on effects
around the Oregon rate cap. Journal of Banking and Finance 34, 546-556.
Appendix
A. Proof
We employ conditional probability premised on Bayes‟ rule with notations B: Borrow and
NB: No Borrow, respectively to derive at:
(i)
nt
ntnt
ntntNBP
NBNBPNBNBP
1
1
nt
ntnt
NBP
BNBP
1 , where is the universal set
nt
ntntnt
NBP
BNBPNBP
11
nt
ntntnt
NBP
BNBPNBP
11
nt
ntntnt
1
11 11
Subsequent reiterations yield the following in steady state:
1
11
1
21
(ii) ntntntnt NBNBPNBBP 11 1
Subsequent reiterations yield the following in the steady state:
1
211
1
1
1
211
B. Figures
Figure 1. Members financial conduct with default outcomes under each model.
Figure 2. All members are savers and also liquidity constrained borrowers.
Figure 3. A fraction of liquidity constrained members borrow from the circuit ‘λ’ and
supported by ‘1-λ’ non-borrowing members.
C(1-ζ) C(1-ζ) C(1-ζ) -Q(1-ζ) Q(1-α)+S(2α-1)
-m -m+2 -m+1 1-n 1
C(1-ζ) C(1-ζ) C(1-ζ) -λQ(1-ζ) λQ(1-α)+S(2λα-1)
-m -m+2 -m+1 1-n 1
Figure 4. A fraction of liquidity constrained members borrow from the circuit ‘λ’
across multi-periods and this reaches a steady state.
Figure 5. Effect of transaction cost on the circuit potential maximum loan.
1-λt-n
1-ρt+1-n λt-n
ρt+1-n
C(1-ζ) C(1-ζ) C(1-ζ) -λt-nQ(1-ζ) -λt+1-nQ(1-ζ)
-m -m+2 -m+1 t-n t+1-n
-λT-1Q(1-ζ)
... T-1 T
λTQ(1-α)
+S(2λTα-1)
1-φt+1-n
φt+1-n
C. Tables
Table 1. Indicative pricing structure of the endogenous interest-free payday loan circuit
Increase in exogenous factor Direction of change in endogenous factor
Cmax Smax Qmax
y + + +
b1 + + +
b2 uc uc +
r uc uc +
ζ uc uc –
α – – –
m + + +
n uc uc –
λ uc uc –
T uc uc +
Notes: Direction of change in the endogenous values „+‟ increase; „–‟decrease, and „uc‟ unchanged.
Table 2. Results illustrating circuit Pareto-efficiency for each model
Model 1
b1 b2 r ζ α m n Cmax Smax Qmax1 QBinding
100 10 12% 2% 10% 6 ¼ 20 160 1,045 200
⅓ 960
15% 2% 10% 6 ¼ 20 160 1,228 200
⅓ 1,113
50 10 12% 2% 10% 6 ¼ 40 320 2,091 200
⅓ 1,921
15% 2% 10% 6 ¼ 40 320 2,456 200
⅓ 2,226
Model 2
b1 b2 r ζ α m n λ Cmax Smax Qmax1 QBinding
100 10 12% 2% 10% 6 ¼ 0.2 20 160 4,044 200
0.3 2,794
0.4 2,170
⅓ 0.2 20 160 3,715 200
0.3 2,567
0.4 1,993
15% 2% 10% 6 ¼ 0.2 20 160 5,026 200
0.3 3,444
0.4 2,652
⅓ 0.2 20 160 4,555 200
0.3 3,121
0.4 2,404
50 10 12% 2% 10% 6 ¼ 0.2 40 320 8,087 200
0.3 5,589
0.4 4,339
⅓ 0.2 40 320 7,430 200
0.3 5,135
0.4 3,987
15% 2% 10% 6 ¼ 0.2 40 320 10,052 200
0.3 6,887
0.4 5,305
⅓ 0.2 40 320 9,109 200
0.3 6,241
0.4 4,807
Model 3
b1 b2 r ζ α m n λ T Cmax Smax Qmax1 Qmax2 QBinding
100 10 12% 2% 10% 6 ¼ 0.2 180 20 3,740 2,417 137 137
0.3 1,611
0.4 1,208
⅓ 0.2 180 20 3,740 2,394 137 137
0.3 1,596
0.4 1,197
¼ 0.2 240 20 4,940 2,417 137 137
0.3 1,611
0.4 1,208
⅓ 0.2 240 20 4,940 2,394 137 137
0.3 1,596
0.4 1,197
15% 2% 10% 6 ¼ 0.2 180 20 3,740 2,279 137 137
0.3 1,520
0.4 1,140
⅓ 0.2 180 20 3,740 2,253 137 137
0.3 1,502
0.4 1,127
¼ 0.2 240 20 4,940 2,279 137 137
0.3 1,520
0.4 1,140
⅓ 0.2 240 20 4,940 2,253 137 137
0.3 1,502
0.4 1,127
50 10 12% 2% 10% 6 ¼ 0.2 180 40 7,480 4,833 274 200
0.3 3,222
0.4 2,417
⅓ 0.2 180 40 7,480 4,788 274 200
0.3 3,192
0.4 2,394
¼ 0.2 240 40 9,880 4,833 274 200
0.3 3,222
0.4 2,417
⅓ 0.2 240 40 9,880 4,788 274 200
0.3 3,192
0.4 2,394
15% 2% 10% 6 ¼ 0.2 180 40 7,480 4,559 274 200
0.3 3,039
0.4 2,279
⅓ 0.2 180 40 7,480 4,506 274 200
0.3 3,004
0.4 2,253
¼ 0.2 240 40 9,880 4,559 274 200
0.3 3,039
0.4 2,279
⅓ 0.2 240 40 9,880 4,506 274 200
0.3 3,004
0.4 2,253
Notes: The model is solved for endogenous variables Cmax, Smax and Qmax where Cmax is the maximum monthly contribution,
Smax is the maximum savings accumulated from the contributions, and Qmax is the potential maximum loan per period. Total
loan advanced is given by QBinding = min {y/b2, Qmax}. The values of the endogenous variables depicted in the table above
signify the breakeven threshold that ensures the circuit is Pareto-neutral. For this simulation, the exogenous parameters are:
(i) member monthly after tax income: y = $2,000; (ii) income multiplier constraint: b1,1 = 80 times, b1,2 = 40 times; (iii) loan
multiplier constraint: b2 = 25 times; (iv) cost of funds: r1 = 12%, r2 = 15%; (v) transaction and default costs: ζ = 2% and α
=10%; (vi) membership gestation period: m = 6 months; (vii) fraction of borrowers in the circuit: λ1 = 0.2, λ2 = 0.3; λ3 = 0.4;
(viii) loan tenure: n1 = 7 days (¼ month); n2 = 10 days (⅓ month); (ix) loan commencement period: t = 1; and (x) circuit life:
T1 = 15 years (180 months), T2 = 20 years (240 months).
Glossary of Arabic terms
Arabic term Closest English meaning
fiqh Islamic jurisprudence.
hiba Voluntary gift.
ijtihād Literally „exertion‟. It implies independent deduction of laws not
self-evident from the primary sources, namely the Qur‟ān and
Sunnah.
qard The act of extending interest-free loan from one‟s property without
expectations of gains. It is also known as qard hasan or qardah-yi
hasanah.
Qur‟ān The holy book of Islam.
ribā An injunction protecting property rights. This is generally
misinterpreted as usury or interest.
ribā an-nasi‟ah This is termed as evident ribā. It is generally an injunction to deter
expropriation of assets on deferred exchanges. It also mitigates
financial fragility and the exclusion of underprivileged from
financial services.
sadaqah Voluntary offering or alms from a person‟s wealth.
salaf The granting of loan subjected to repayment at a later time.
shari‟ah Islamic law
sunnah The practices of the Prophet Muhammad PBUH. Along with the
Qur‟ān and Hadith (recorded sayings of the Prophet Muhammad),
it is a major source of Shari‟ah, or Islamic law.
takāful Islamic insurance that functions similar to mutuals.
waqf (awqāf –
plural)
Philanthropic foundations involving the endowment of privately
owned properties for charitable purposes in perpetuity.
zakāt Literally „cleansing or purity‟. It implies a religious tax to be
deducted from one‟s wealth to help the needy.